GlobColour Product User Guide. Ref: GC-UM-ACR-PUG-01 Date : 13/02/2015 Issue : version 3.1 Page : 1 GLOBCOLOUR

(1)

GLOBCOLOUR

Product User’s Guide

Reference: GC-UM-ACR-PUG-01

Version 3.0

(2)

Document Signature Table

Name Function Company Signature Date

Author ACRI-ST GlobColour Team

ACRI-ST

Verification A. Mangin Scientific Director

ACRI-ST

Approval O. Fanton d’Andon President ACRI-ST

Change record

Issue Date Change Log Version

1.0 17/11/06 Product User Guide Initial version

1.1 22/12/06 Updated pages version 1.1

2.0 08/01/10 New parameters have been added (to be compliant with MKL3

outputs)

version 2.0

3.0 draft 27/10/14 GlobColour 2nd reprocessing Draft version

version 3.0 draft

3.0 5/12/14 GlobColour 2nd reprocessing version 3.0

(3)

Table of content

1 INTRODUCTION ... 7

1.1 Background ... 7

1.2 Scope of the document ... 7

1.3 Acronyms ... 8

1.4 Brief overview of GlobColour products ... 9

1.4.1 Parameters ... 9

1.4.2 Spatial Domain ...10

1.4.3 Sensors ...11

1.4.4 Spatial and Temporal resolutions ...11

2 THE PRODUCTS CONTENT ... 12

2.1 Parameters overview ... 12

2.1.1 Biological parameters ...12

2.1.2 Atmospheric Optical parameters ...13

2.1.3 Ocean Surface Optical parameters ...13

2.1.4 Ocean Subsurface Optical parameters...14

2.1.5 OSS2015 Demonstration Biological Products ...15

2.2 Parameter Detailed Description ... 15

2.2.1 CHL1 ...16 2.2.2 CHL2 ...18 2.2.3 TSM ...19 2.2.4 PIC ...20 2.2.5 POC ...21 2.2.6 NFLH ...22 2.2.7 WVCS ...23 2.2.8 Txxx ...24 2.2.9 Axxx ...26 2.2.10 CF ...28 2.2.11 ABSD ...29 2.2.12 (N)RRSxxx ...30 2.2.13 EL555...32 2.2.14 PAR ...33 2.2.15 BBP...34 2.2.16 CDM...35 2.2.17 KD490 ...37 2.2.18 KDPAR ...39 2.2.19 ZHL ...40 2.2.20 ZEU...41 2.2.21 ZSD...42 2.2.22 CHL-CIA ...44 2.2.23 BBPxxx-LOG ...45 2.2.24 BBPS-LOG ...46 2.2.25 PSD-XXX ...47 2.2.26 POC-SURF...48

(4)

2.2.29 PHYSAT ...52

2.3 The spatial and temporal coverage ... 53

2.3.1 The binned products ...53

2.3.2 The mapped products ...53

3 THE GLOBCOLOUR SYSTEM ... 54

3.1 Overall description of the processor ... 54

3.2 The preprocessor... 56

3.2.1 MERIS...57

3.2.2 MODIS/SeaWiFS/VIIRS ...59

3.3 The spatial and temporal binning schemes ... 60

3.4 The GlobColour data-day approach ... 60

4 THE PRODUCTS FORMAT ... 65

4.1 General rules ... 65

4.2 Naming convention ... 65

4.3 The binned products ... 66

4.4 The mapped products ... 72

5 HOW TO…? ... 74

5.1 Access the GlobColour data ... 74

5.1.1 The HERMES interface ...74

5.1.2 Ordering GlobColour Products ...75

5.1.3 Ordering OSS2015 demonstration products ...78

5.1.4 Ordering a list of products ...78

5.1.5 Retrieving the data ...78

5.2 Download the data from the GlobColour ftp server ... 79

5.3 Read the data ... 79

5.4 Visualize the data ... 80

6 APPENDICES ... 83

6.1 Global ISIN grid definition ... 83

6.2 Summary of products content ... 84

6.3 The main characteristics of the products ... 85

6.4 The steps of the binning and merging schemes... 86

6.4.1 Step 1: L2 to L3 track ...86

6.4.2 Step 2: L3 track to L3 daily for each single instrument ...87

6.4.3 Step 3: L3 daily for each single instrument to merged L3 daily ...88

6.4.4 Step 4: L3 daily merged to 8-days and monthly L3 ...89

6.4.5 Step 5: L3 daily/8days/monthly merged products to mapped products ...89

6.4.6 Step 6: generation of the quicklooks ...90

6.5 The error bars... 90

6.6 Common Data Language description ... 91

6.7 References ... 94

6.7.1 GlobColour products references ...94

(5)

List of Tables

Table 1-1: Available sensors in the GlobColour data set ...11

Table 1-2: Overview of the GlobColour products ...11

Table 2-1: List of GlobColour Biological Parameters ...12

Table 2-2: List of GlobColour Atmospheric Optical parameters ...13

Table 2-3: List of GlobColour Ocean Surface Optical parameters ...13

Table 2-4: GlobColour Ocean Subsurface parameters ...14

Table 2-5: Main characteristics of the ISIN grids ...53

Table 3-1: List of variables with a specific preprocessing...56

Table 3-2: List of parameters and filters applied to the MERIS level 2 data ...58

Table 3-3: List of parameters and filters applied to the MODIS/SeaWiFS/VIIRS level 2 data 59 Table 3-4: Input parameters for data-day classification...64

Table 3-5: CNT of satellites...64

Table 4-1: Dimensions - binned products...67

Table 4-2: Variables - binned products ...67

Table 4-3: Flags description ...68

Table 4-4: Variables attributes - binned products ...69

Table 4-5: Global attributes - binned products (1/3) ...70

Table 4-8: Dimensions of the grid - mapped products ...72

Table 4-9: Dimensions - mapped products ...73

Table 4-10: Variables - mapped products (1/2) ...73

Table 4-11: Variables - mapped products (2/2) ...73

Table 6-1: ISIN grid definition ...83

Table 6-2: Summary of products content ...84

(6)

List of Figures

Figure 1-1: A sample GlobColour Global product ...10

Figure 1-2: A sample GlobColour Europe product ...10

Figure 3-1: The GlobColour processor high-level description ...55

Figure 3-2: MERIS pixels UTC as a function of the pixel longitude (35 days - October 2003) ...61

Figure 3-3: MODIS pixels UTC as a function of the pixel longitude (1 day - June 2003) ...62

Figure 3-4:SeaWiFS pixels UTC as a function of the pixel longitude (1 day - December 2003) ...62

Figure 3-5: Data-day definition line above MODIS pixels UTC versus longitude plot. ...63

Figure 3-6: Data-day definition line above one SeaWiFS track. ...63

Figure 5-1: the HERMES Interface ...74

Figure 5-2: GlobColour Data Access interface ...75

Figure 5-3: Selection of the map overlays ...76

Figure 5-4: Selected products list and previsualisation screen ...77

Figure 5-5: OS2015 products data access ...78

Figure 5-6: Visualization of a GlobColour L3m product using Visat ...80

Figure 5-7: Visualization of a GlobColour L3m product using ncview. ...80

Figure 5-8: Visualizing GlobColour products with matplotlib...81

(7)

1

Introduction

1.1

Background

The GlobColour project started in 2005 as an ESA Data User Element (DUE) project to provide a continuous data set of merged L3 Ocean Colour products. Merging outputs from different sensors ensures data continuity, improves spatial and temporal coverage and reduces data noise. This allows in particular to process long time series of consistent products (trend analysis, climatology, data assimilation for model hindcast).

Since then, ACRI has maintained the archive and Near Real Time data access services through the Hermes website; knowing that in the frame of MyOcean, a sub-set (Chl, reflectances, Secchi depth) is produced at global scale and in Near Real Time by the OCTAC Processing Unit.

The 20014 reprocessing and the update of the Hermes interface were performed in the framework of the OSS2015 project, with funding from the EU FP7 under grant n°282723. The GlobColour project has received additional funding from European Union FP7 under grant agreement n° 218812 (MyOcean) and from PACA Region under project RegiColour In addition, support from NASA regarding access to L2 products is acknowledged.

The GlobColour primary data set has now been delivered as a Group on Earth Observations System of Systems (GEOSS) core data set under reference:

urn:geoss:csr:resource:urn:uuid:4e33fd81-d5cc-dc40-b645-ab961447d9d8.

1.2

Scope of the document

This User Guide contains a description of:  the products content

- the parameters

- the spatial and temporal coverage - the processing system

 the products format

 Hermes interface user’s guide

(8)

1.3

Acronyms

AV Simple average method

AVW Weighted average method

bbp Particulate back-scattering coefficient

BEAM Basic ERS and Envisat (A)ATSR and MERIS Toolbox BOUSSOLE Bouée pour l’acquisition de Séries Optiques à Long Terme

CDL Common Data Language

CDM Coloured dissolved and detrital organic materials absorption coefficient

CF Climate and Forecast

CF Cloud Fraction

CHL Chlorophyll-a

CZCS Coastal Zone Color Scanner

DLR Deutsches Zentrum für Luft- und Raumfahrt

DPM Detailed Processing Model

DUE Data User Element of the ESA Earth Observation Envelope Programme II

EEA European Environment Agency

EL555 Relative excess of radiance at 555 nm (%)

EO Earth observation

GHRSST-PP GODAE High Resolution Sea Surface Temperature - Pilot Project GSM Garver, Siegel, Maritorena Model

ICESS Institute for Computational and Earth Systems Science IOCCG International Ocean Colour Coordinating Group

IOCCP International Ocean Carbon Coordination Project IODD Input Output Data Definition

ISIN Integerised SINusoidal projection

LOV Laboratoire Océanologique de Villefranche-sur-mer

LUT Look-Up Table

MER Acronym for the MERIS instrument used in the GlobColour filenames MERIS Medium Resolution Imaging Spectrometer

MERSEA Marine Environment and Security for the European Area – Integrated Project of the EC Framework Programme 6

MOBY Marine Optical Buoy

MOD Acronym for the MODIS instrument used in the GlobColour filenames MODIS Moderate Resolution Imaging Spectrometer

netCDF Network Common Data Format

NIVA Norwegian Institute for Water Research

(N)RRSXXX Fully normalised remote sensing reflectances at xxx nm (sr-1)

NRT Near-real time

(9)

PC Plate-Carré projection

PNG Portable Network Graphics

RD Reference Document

ROI / RoI Region of Interest

SeaBASS SeaWiFS Bio-Optical Archive and Storage System SeaWiFS Sea-Viewing Wide Field of View Sensor

SAA Sun Azimuth Angle

SWF Acronym for the SeaWiFS instrument used in the GlobColour filenames

SZA Sun Zenith Angle

TSM Total Suspended Matter

T865 Aerosol optical thickness over water

UCAR University Corporation for Atmospheric Research UoP University of Plymouth (U.K)

UTC Coordinated Universal Time

VAA Viewing Azimuth Angle

VZA Viewing Zenith Angle

1.4

Brief overview of GlobColour products

1.4.1 Parameters

The parameters of the GlobColour data set are:

 Biological parameters: Chlorophyll (several algorithms), Particulate Organic/Inorganic Carbon, Fluorescence…

 Atmosphere optical parameters: aerosol thickness, cloud fraction, water vapour column…

 Ocean-surface optical parameters: reflectances

 Sub-surface optical parameters: attenuation and back-scattering coefficients, turbidity The full list of parameters is provided in section 2.1.

For some parameters, several alternative algorithms are proposed. This is the case when no algorithm is clearly superior to the other(s). The relative performance may vary depending on the conditions (water types, regions, sensors…), and users are advised to compare the results on a case-by-case basis.

(10)

1.4.2 Spatial Domain

Two spatial domains are covered:

 the global Earth domain:

Figure 1-1: A sample GlobColour Global product  an extended Europe area at full resolution (1km):

(11)

1.4.3 Sensors

The GlobColour data set is built using the following sensors:

Sensor Product type Start Date End Date Reprocessing

MERIS RR 1 km March 2002 April 2012 ESA 3rd reprocessing 2011

MODIS AQUA 1 km July 2002 present NASA R2013.1

SeaWIFS GAC 4 km Sept. 1997 Dec. 2010 NASA R2010.0

VIIRS 1 km Oct. 2011 Dec. 2014 NASA R2013.1

VIIRS 1 km Dec. 2014 Present NASA R2014.0

Table 1-1: Available sensors in the GlobColour data set

The data set includes single-sensor and merged products. Merged products are generated for three merging techniques:

 simple averaging  weighted averaging  GSM model

The relative performance of the weighting methods depends on the conditions (water types, region, glint/aerosol conditions…) Users are advised to compare the results on a case-by-case basis as far as possible.

Note: The GlobColour reprocessing R2014 archive uses two different VIIRS reprocessing (R2013.1 before December 2014 and R2014.0 after). This may create discontinuities in the VIIRS time series. This problem will be fixed in the next GlobColour reprocessing.

1.4.4 Spatial and Temporal resolutions

The spatial and temporal resolutions of the products distributed to the end-users are:

Spatial domain Grid Temporal domain Resolution

Global ISIN Daily, 8 days, monthly 1/24°

Europe ISIN Daily, 8 days, monthly 1/96°

Global PC Daily, 8 days, monthly 1/24°, 0.25°, 1.0°

Europe PC Daily, 8 days, monthly 0.01°

Table 1-2: Overview of the GlobColour products

Note: SeaWIFS GAC products have a spatial sampling distance of 4 km approximately.

(12)

2

The products content

2.1

Parameters overview

This section provides the detailed description of the exhaustive list of all parameters that are available in the GlobColour products.

The GlobColour merged products are generated by different simple averaging techniques (see IOCCG reports N°4 and 5) or by the use of the GSM model (see Maritorena and Siegel, 2005).

In the following tables, the following acronyms are used:

AV: simple averaging, AVW: weighted averaging, GSM: GSM model, AN: analytical from other L3 products, STAT: classification statistics.

2.1.1 Biological parameters

The GlobColour biological Parameters are listed in Table 2-1.

Parameter Description L3 merging method

Sensor availability

MER MOD SWF VIR

CHL1 Chlorophyll concentration (mg/m 3_{) – GSM} algorithm AVW, GSM     CHL2 Chlorophyll concentration (mg/m 3_{) – Neural} Network algorithm AV 

TSM Total suspended matter concentration (g/m3) AV 

PIC Particulate Inorganic Carbon (mol/m3₎ _AVW _ _ _

POC Particulate Organic Carbon (mg/m3) AVW   

NFLH Normalised Fluorescence Line Height

(mW/cm²/microm/sr) AV 

(13)

2.1.2 Atmospheric Optical parameters

Parameter Description L3 merging method

MER MOD SWF VIR

WVCS Total water vapor column, clear sky (g/cm²) AV 

T865 Aerosol optical thickness over water (-) AVW    

A865 Angstrom alpha coefficient over water (-) AVW    

T443 Aerosol optical thickness over land (-) AV 

A443 Angstrom alpha coefficient over land (-) AV 

T550 Aerosol optical thickness over water+land (-) AN    

A550 Angstrom alpha coefficient over water+land (-) AN    

CF Cloud fraction (%) STAT    

ABSD ABSOA_DUST flag statistics (%) STAT 

Table 2-2: List of GlobColour Atmospheric Optical parameters

2.1.3 Ocean Surface Optical parameters

Parameter Description L3 merging

method

MER MOD SWF VIR

NRRS412

Fully normalised remote sensing reflectance at xxx nm (sr-1) AVW     NRRS443 AVW     NRRS469 AV  NRRS490 AVW     NRRS510 AVW   NRRS531 AV  NRRS547 AV  NRRS551 AV  NRRS555 AVW (1)     NRRS560 AV  NRRS620 AV  NRRS645 AV  NRRS670 AVW     NRRS678 AV 

RRS681 Non normalised remote sensing reflectance at xxx nm (sr-1)

AV 

RRS709 AV 

EL555 Relative excess of radiance at 555 nm (%) AN    

PAR Photosynthetically Available Radiation

(14)

(1): spectral inter-calibration is applied prior to the merging.

2.1.4 Ocean Subsurface Optical parameters

Parameter Description L3 merging method

MER MOD SWF VIR

BBP Particulate back-scattering coefficient at

443 nm (m-1) GSM    

CDM Coloured dissolved and detrital organic

materials absorption coefficient at 443 nm (m-1)

AVW  

AV (1) 

GSM    

KD490 Diffuse attenuation coefficient at 490 nm (m -1

)

Algorithms of Morel and Lee AN (2 methods)    

KDPAR

Diffuse attenuation coefficient for the Photosynthetically Available Radiation (m-1) Algorithms of Morel

AN    

ZHL Heated layer depth (m) AN    

ZEU Depth of the bottom of the euphotic layer (m) AN    

ZSD Secchi disk depth (m)

Algorithms of Morel and Doron AN (2 methods)    

Table 2-4: GlobColour Ocean Subsurface parameters

(1): MERIS CDM is not mergeable with MODIS and SeaWiFS because corresponding L2 algorithms are different.

(15)

2.1.5 OSS2015 Demonstration Biological Products

This section lists the demonstration products developed in the frame of the OSS2015 project. All the products are available as merged products (averaging method) at global scale with 25 km resolution.

Archive and product format conventions are not applicable to Third-Party products PP and PHYSAT.

Parameter Description Product type

BBP-xxx-LOG Particulate Back-scattering coefficients at xxx nm (m-1) - LOG algorithm

Monthly analytical

BBPS Spectral slope of the particulate back-scattering coefficient (-) Monthly analytical POC-SURF Surface Particulate Organic Carbon Concentration (mg/m3) Monthly analytical POC-INT Column-integrated Particulate Organic Carbon Concentration

(mg/m2)

Monthly analytical

PSD-XXX Number concentration of pico, nano and micro particles (#/m3) Monthly analytical PP-AM Primary Production - Antoine- Morel Algorithm Third party,

monthly analytical PP-UITZ Primary Production - Uitz Algorithm Third party,

monthly analytical PHYSAT Phytoplankton Functional Types Third party,

monthly analytical

(16)

2.2.1 CHL

1

Category Biological PARAMETER CHL1

Description L3 merging method Sensor availability

Chlorophyll concentration (mg/m3) for case 1 waters (see below). It is commonly used as a proxy for the biomass of the phytoplankton.

AV, AVW, GSM MER MOD SWF VIR

AVW

(17)

Algorithm _Reference

SWF: OC4v5 0’Reily et al., 2000

MER: OC4Me 0’Reily et al., 2000

MOD/VIR: OC3v5 0’Reily et al., 2000 AVW: weighted average of single-sensor Level 2 CHL1 products -

GSM: GSM merging of single sensor L3 NRRS. The GSM method uses the normalized reflectances at the original sensor wavelengths, without intercalibration. CHL1 is one of the outputs of the method, in addition to bbp and CDM.

Maritorena and Siegel, 2005

Validity

The CHL1 algorithms are applicable for “case 1” waters, i.e. waters where the phytoplankton concentration dominates over inorganic particles.

References

O'Reilly, J.E., and 24 Coauthors, 2000: SeaWiFS Postlaunch Calibration and Validation Analyses, Part 3. NASA Tech. Memo. 2000-206892, Vol. 11, S.B. Hooker and E.R. Firestone, Eds., NASA Goddard Space Flight Center. http://oceancolor.gsfc.nasa.gov/REPROCESSING/R2009/ocv6/

Maritorena, S. and Siegel, D.A. 2005. Consistent Merging of Satellite Ocean Colour Data Sets Using a Bio-Optical Model. Remote Sensing of Environment, 94, 4, 429-440.

Maritorena S., O. Hembise Fanton d’Andon, A. Mangin, and D.A. Siegel. 2010. Merged Satellite Ocean Color Data Products Using a Bio-Optical Model: Characteristics, Benefits and Issues. Remote Sensing of Environment, 114, 8: 1791-1804.

Related products

Other Chlorophyll-a products: CHL2, CHL-CIA (OSS2015 demonstration product) Other GSM algorithm outputs: CDM, bbp

(18)

2.2.2 CHL

2

Category Biological PARAMETER _CHL2

CHL2 is the chlorophyll concentration (mg/m3) for

Case 2 waters (see section validity). It is commonly used as a proxy for the biomass of the

phytoplankton.

AV MER

Algorithm Reference

CHL2 uses the MERIS C2R Neural Network algorithm. Another

output of the algorithm is TSM. Doerffer and Schiller (2007)

Validity

The product is valid for case 2 waters, i.e. waters where inorganic particles dominate over phytoplankton (typically in coastal waters).

References

The MERIS Case 2 water algorithm, R. Doerffer, H. Schiller, International Journal of Remote Sensing, Vol. 28, Iss. 3-4, 2007, doi:10.1080/01431160600821127

Related products

Other Chlorophyll-a products: CHL1, CHL-CIA (OSS2015 demonstration product) Other C2R algorithm output: TSM

(19)

2.2.3 TSM

Category Biological PARAMETER _TSM

TSM is the total suspended matter concentration

(g/m3). It is a measure of the turbidity of the water. AV MER

TSM uses the MERIS C2R Neural Network algorithm. Another output of the algorithm is CHL2.

Doerffer and Schiller (2007)

Validity

The product is valid for case 2 waters, i.e. waters where inorganic particles dominate over phytoplankton (typically in coastal waters).

References

Related products

The MERIS TSM product is computed from the back-scattering coefficient at 444 nm using the following assumptions: BP = BBP/0.015, TSM = 1.73*BP. Therefore the BBP variable issued from the GSM algorithm is closely related to TSM.

(20)

2.2.4 PIC

Category Biological PARAMETER PIC

Description L3 merging

method Sensor availability

PIC is the Particulate Inorganic Carbon or suspended calcium carbonate concentration (mol/m3). CaCO3 is produced in shallow waters by either coral reefs or macrophytic algae, or in the plankton, by coccolithophores, foraminifera, and pteropods.

AV, AVW SWF MOD VIR

Algorithm Reference

PIC uses the original NASA algorithms: 2-band look-up table approach and 3-band algorithm at high concentrations

Bach et al. (2005) Gordon et al. (2001)

References

Balch, W. M., H. R. Gordon, B. C. Bowler, D. T. Drapeau, and E. S. Booth. (2005) Calcium carbonate measurements in the surface global ocean based on Moderate-Resolution Imaging

Spectroradiometer data, JGR, Vol. 110, C07001 http://dx.doi.org/10.1029/2004JC002560

Gordon, Howard R., G. Chris Boynton, William M. Balch, Stephen B. Groom, Derek S. Harbour, and Tim J. Smyth. (2001) Retrieval of Coccolithophore Calcite Concentration from SeaWiFS Imagery. GRL, Vol. 28, No. 8, pp. 1587-1590 http://dx.doi.org/10.1029/2000GL012025

Related products

(21)

2.2.5 POC

Category Biological PARAMETER _POC

POC is the Particulate Organic Carbon (mol/m3). POC is an important component in the carbon cycle and serves as a primary food sources for aquatic food webs.

AV, AVW SWF MOD VIR

POC uses the original NASA algorithm (correlation of band ratios). Stramski et al. (2008)

References

Stramski. D., R.A. Reynolds, M. Babin, S. Kaczmarek, M.R. Lewis, R. Rottgers, A. Sciandra, M. Stramska, M.S. Twardowski, B.A. Franz, and H. Claustre (2008). Relationships between the surface concentration of particulate organic carbon and optical properties in the eastern South Pacific and eastern Atlantic Oceans, Biogeosci., 5, 171-201.

PIC provides the Particulate Inorganic Concentration

(22)

2.2.6 NFLH

Category Biological PARAMETER POC

NFLH is the Normalised Fluorescence Line Height

(mW/cm²/microm/sr) at 678 nm. Fluorescence is a marker of the photosynthetic activity of the phytoplankton.

AV MOD

NFLH is based on the original MODIS L2 product using the spectral

band of MODIS at 678 nm. Berhenfeld et al. (2009)

References

Behrenfeld, M.J., T.K. Westberry, E.S. Boss, R.T. O'Malley, D.A. Siegel, J.D. Wiggert, B.A. Franz, C.R. McClain, G.C. Feldman, S.C. Doney, J.K. Moore, G. Dall'Olmo, A. J. Milligan, I. Lima, and N. Mahowald (2009). Satellite-detected fluorescence reveals global physiology of ocean phytoplankton, Biogeosci., 6, 779-794.

(23)

2.2.7 WVCS

Category Atmosphere PARAMETER _WVCS

method Sensor availability

WVCS is the total water vapor column over clear sky (g/cm²) coming from the MERIS L2 data. Water vapour is the most effective greenhouse gas in the atmosphere. It influences weather and climate and is responsible for cloud

development, precipitation, and modulates the atmospheric radiative energy transfer.

AV MER

WCVS is an average of the corresponding MERIS product which uses a

polynomial function of several band ratios. MERIS ATBD

References

(24)

2.2.8 Txxx

Category Atmosphere PARAMETER _Txxx

Txxx (xxx=443, 550, 865) are the aerosol optical thicknesses at 443 (over land), 550 (over land and water) and 865 nm (over water). The optical thickness is the logarithm of the ratio between the down-welling irradiances and the bottom of the atmosphere. 443: AV 550: AN 865: AVW MER MER MER MOD MOD SWF SWF VIR VIR T443 (Land) T550 (Land + Water)

(25)

T865 (Water)

Algorithm Reference

T443 is computed from the MERIS L2 product.

T865 is merged from corresponding L2 products for the various sensors. T550 is extrapolated from the land and water products using the

corresponding Angström  exponent: T550 = Txxx * (550 / xxx)-

with =1 for land (xxx=443) and =A865 for water (xxx=865)

Gordon, 1997 MERIS ATBD

Validity

The GlobColour merged atmosphere products are not yet validated. The validity of the products is not certified.

References

Gordon, H.R., 1997. Atmospheric correction of ocean color imagery in the Earth Observing System era. Journal of Geophysical Research —Atmospheres 102 (D14), 17081–17106

(26)

2.2.9 Axxx

Category Atmosphere PARAMETER _Axxx

Axxx (xxx=443, 550, 865) are the Angstöm exponents at 443 (over land), 550 (over land and water) and 865 nm (over water). The Angström  exponent is related to the aerosol optical thickness  by 𝜏1⁄𝜏0= (𝜆1⁄ )𝜆0 −𝛼. 443: AV 550: AN 865: AVW MER MER MER MOD MOD SWF SWF VIR VIR A443 (Land)

(27)

A865 (Water)

Algorithm Reference

A443 is computed from the MERIS L2 product.

A865 is a weighted average of the corresponding products from all sensors. A550 is obtained by merging A443 (over land) and A865 (over water) products.

MODIS ATBD MERIS ATBD

Validity

The GlobColour merged atmosphere products are not yet validated. The validity of the products is not certified.

References

MODIS ATBD : http://modis-atmos.gsfc.nasa.gov/MOD06_L2/atbd.html (accessed December 2014)

(28)

2.2.10 CF

Category Atmosphere PARAMETER CF

CF is the Cloud Fraction (%), i.e. the percentage of pixels

with flags Cloud, Ice, or haze per bin. STAT MER MOD SWF VIR

This parameter is determined by using the following flags:

 CLDICE = "Probable cloud or ice contamination" for MODIS, SeaWiFS and VIIRS instruments

 "CLOUD or (WATER and ICE_HAZE)" where ICE_HAZE = "Ice or high aerosol load pixel or Cloud" for the MERIS

instrument. Two products are available:

 daily products: percentage of input pixels per bin flagged as cloudy in the original level 2 products

 8-days and monthly products: percentage of merged days per bin where the daily cloud fraction is greater than a specified threshold (50% in current GlobColour processor)

MODIS ATBD MERIS ATBD

References

MODIS ATBD : http://modis-atmos.gsfc.nasa.gov/MOD06_L2/atbd.html (accessed December 2014) MERIS ATBD: https://earth.esa.int/handbooks/meris/CNTR2-7.htm (accessed December 2014)

(29)

2.2.11 ABSD

Category Atmosphere PARAMETER _ABSD

ABSD is the MERIS L2 ABSOA_DUST flag statistics (%), which is raised to indicate the presence of dust-like absorbing aerosols.

STAT MER

ABSD is computed from the corresponding MERIS L2 product. MERIS ATBD

References

(30)

2.2.12 (N)RRSxxx

Category Ocean Surface Optical PARAMETER (N)RRSxxx

The (N)RRSxxx are the remote sensing reflectances at xxx nm (expressed in sr-1). The remote sensing reflectance is the ratio of the upwelling radiance to the downwelling irradiance at Sea surface.

NRRS reflectances are fully normalized. RRSS681 and RRS760 are not normalized.

(1) For NRRS 555, an inter-calibration is performed before merging (see algorithm description below).

412 AVW MER MOD SWF VIR 443 AVW MER MOD SWF VIR 469 AV MOD 490 AVW MER MOD SWF VIR 510 AVW MER SWF 531 AV MOD 547 AV MOD 551 AV VIR 555 AVW

(1) MER MOD SWF VIR 560 AV MER 620 AV MER 645 AV MOD 670 AVW MER MOD SWF VIR 678 AV MOD 681 AV MER 709 AV MER

(31)

Algorithm

MERIS normalised water leaving reflectances from the L2 products are initially converted into fully normalised water leaving reflectances (except for the MERIS 681 nm fluorescence band and for the 709 nm band).

The NRRSxxx daily L3 products are generated for each instrument, using the corresponding L2 data. The merged NRRSxxx concentration is then computed as the weighted average of all the single-sensor products.

The 547-560 nm bands are submitted to a specific processing just before averaging to prepare a more consistent merging between the instruments. First of all, all bands are spectrally re-affected to 555 nm, using an inter-spectral conversion LUT which is a function of the CHL1 concentration (weighted average version):

 MODIS: NRRS555 = NRRS547 * (0.93573 + 0.0861 * y + 0.01545 * y2 - 0.00714 * y3 - 0.00245 * y4)

 VIIRS: NRRS555 = NRRS551 * (0.97979 + 0.03583 * y + 0.0057 * y2 - 0.00277 * y3 - 0.00085 * y4)

 SeaWiFS: No change as SeaWiFS band is actually at 555 nm

 MERIS: NRRS555 = NRRS560 * (1.02542 - 0.03757 * y - 0.00171 * y2 + 0.0035 * y3 + 0.00057 * y4),

where y = log10(CHL1).

Validity

The validity limit for the spectral interpolation method at 555 nm is 0.01 ≤ CHL1 ≤ 30.

References

N/A

(32)

2.2.13 EL555

Category Ocean Surface Optical PARAMETER _EL555

method Sensor availability EL555 is an indicator of an excess of luminance at 555 nm (%)

after removal of the chlorophyll contribution in case 1 water. It is an indicator of the quality of the Chlorophyll retrieval and may indicate that the presence of other constituents (especially suspended matter) might have impact the inversion.

AN MER MOD SWF VIR

The parameter is computed from the corresponding merged fully normalised remote sensing reflectance at 555 nm and the CHL1 (weighted method) products, using the following algorithm:

If (CHL1 > 0.2) and (NRRS555 > Rholim(CHL1)) then

raise the turbid flag for all products

EL555 = 100. [ NRRS555 – Rholim(CHL1) ] / Rholim(CHL1)

Endif,

where Rholim(CHL1) is expressed as:

               3 2 lim 10 y 0006382 . 0 y 00099233 . 0 y 006665 . 0 0104 . 0 y Rho 1 CHL log y

Morel and Bélanger 2006

References

Morel, A. and S. Bélanger, (2006) Improved Detection of turbid waters from Ocean Color information,

(33)

2.2.14 PAR

Category Ocean Subsurface Optical PARAMETER _PAR

PAR is the Photosynthetically Available Radiation

(einstein/m²/day). It is the mean daily photon flux density in the visible range (400 to 700 nm) that can be used for photosynthesis.

AV, AVW MOD SWF VIR

PAR uses the original L2 products. Frouin et al.

References

http://oceancolor.gsfc.nasa.gov/DOCS/seawifs_par_wfigs.pdf

Frouin, R., B. A. Franz, and P. J. Werdell, 2003: The SeaWiFS PAR product. In Algorithm Updates for the Fourth SeaWiFS Data Reprocessing, S. B. Hooker and E. R. Firestone, Editors, CC NASA/TM-2003-206892, Vol. 22, 46-50.

Related products

(34)

2.2.15 BBP

Category Ocean Subsurface Optical PARAMETER _BBP

BBP is the particulate back-scattering coefficient (m-1_{) at the}

reference wavelength of λ0 = 443 nm. The back-scattering

coefficient can be used as a proxy for the concentration of suspended particles in sea water.

GSM MER MOD SWF VIR

BBP is an output of the GSM merging algorithm applied to L3

single-sensors reflectances NRRS. Maritorena and Siegel, 2005

References

Maritorena, S. and Siegel, D.A. 2005. Consistent Merging of Satellite Ocean Colour Data Sets Using a Bio-Optical Model. Remote Sensing of Environment, 94, 4, 429-440.

Related products

The MERIS TSM product is computed from the back-scattering coefficient at 444 nm using the following assumptions: BP = BBP/0.015, TSM = 1.73*BP. Therefore the BBP variable issued from the GSM algorithm is closely related to TSM.

The OSS2015 demonstration products include backscattering coefficients at several wavelength computed by the Non-Spectral Algorithm of Loisel et al. 2006.

(35)

2.2.16 CDM

Category Ocean Subsurface Optical PARAMETER _BBP

Description L3 merging

CDM is the absorption coefficient (m-1) of Coloured Dissolved and detrital organic Materials at the reference wavelength of λ0 = 443 nm. AV AV, AVW GSM MER MER MOD MOD SWF SWF VIR MODIS MERIS

(36)

GSM

SWF/MOD: CDM is derived from the L2 CDOM Index and CHL1 products using the following formula:

CDM = CDOM_index * 0.0316 * CHL10.63.

The L2 CDOM index is based on a band ratio algorithm. Morel and Gentili (2009)

MER: MERIS Neural Network algorithm Doerffer and Schiller (2007) AVW: weighted average of SWF and MOD products. The MERIS product is

not merged because the algorithm is significantly different.

GSM: GSM merging of single sensor L3 NRRS. The GSM method uses the normalized reflectances at the original sensor wavelengths, without intercalibration. CDM is one of the outputs of the method, in addition to BBP and CHL1.

Maritorena et al. 2010

References

Morel, A. and B. Gentili (2009). A simple band ratio technique to quantify the colored dissolved and detrital organic material from ocean color remotely sensed data. Remote Sensing of Environment, 113, 998-1011, doi:10.1016/j.rse.2009.01.008.

Maritorena, S. and Siegel, D.A. 2005. Consistent Merging of Satellite Ocean Colour Data Sets Using a Bio-Optical Model. Remote Sensing of Environment, 94, 4, 429-440.

Related products

Other MERIS Neural Network outputs: CHL2, TSM. Other GSM algorithm outputs: CHL1 (GSM), CDM.

(37)

2.2.17 KD490

Category Ocean Subsurface Optical PARAMETER _{KD490, KD490-LEE}

Kd(490) is the diffuse attenuation coefficient (m-1_{) of the}

downwelling irradiance at 490 nm. It is one indicator of the turbidity of the water column.

KD490 is computed according to the Morel algorithm, while KD490-LEE is computed from the Lee and Arnone algorithm.

AN MER MOD SWF VIR

(38)

KD490-LEE

KD490 is computed from the corresponding merged CHL1 products (weighted method), using the following empirical formula:

KD490 = 0.0166 + 0.08349 * CHL10.63303

A similar formula with slightly different coefficients is described in Morel et al. (2007).

KD490-LEE is computed from the corresponding merged fully normalised remote sensing reflectance products at 443, 490, 555 and 670 nm using the semi-analytical method of Lee and Arnone.

Morel, 2006

Lee and Arnone, 2005

References

André Morel, personnal communication, June 2006.

André Morel, Yannick Huot, Bernard Gentili, P. Jeremy Werdell, Stanford B. Hooker, Bryan A. Franz, “Examining the consistency of products derived from various ocean color sensors in open ocean (Case 1) waters in the perspective of a multi-sensor approach”, Remote Sensing of Environment, 111 (2007), 69-88, doi:10.1016/j.rse.2007.03.012

Zhong-Ping Lee, Ke-Ping Du and Robert Arnone (2005) A model for the diffuse attenuation coefficient of downwelling irradiance. JOURNAL OF GEOPHYSICAL RESEARCH. VOL. 110, C02016, doi:10.1029/2004JC002275, 2005

(39)

2.2.18 KDPAR

Category Ocean Subsurface Optical PARAMETER KDPAR

KdPAR is the diffuse attenuation coefficient (m-1_{) of the}

downwelling Photosynthetically Available Radiation in the 400 to 700 nm range.

AN MER MOD SWF VIR

The merged KDPAR is computed from the corresponding merged KD490 product, using the following equation:

KDPAR = 0.0665 + 0.874 * Kd(490) - 0.00121 / Kd(490)

Morel et al. 2007

References

Morel, A., Huot, Y., Gentili, B., Werdell, P.J., Hooker, S.B. and B.A. Franz (2007). “Examining the consistency of products derived from various ocean color sensors in open ocean (Case 1) waters in the perspective of a multi-sensor approach”. Remote Sensing of Environment, 111, 69-88.

(40)

2.2.19 ZHL

Category Ocean Subsurface Optical PARAMETER _ZHL

Description L3 merging

ZHL is the depth of the bottom of the heated layer (m). AN MER MOD SWF VIR

ZHL is computed from the corresponding merged KDPAR MOREL product, using the following equation:

ZHL = 2 / KDPAR

Morel et al. 2007

References

Related products

(41)

2.2.20 ZEU

Category Ocean Subsurface Optical PARAMETER _ZEU

method Sensor availability ZEU is the depth of the euphotic layer (m), i.e. the depth for

which the down-welling irradiance is 1% of its value at the surface. It characterizes the upper layer of the ocean which can support phytoplankton photosynthesis. It depends on the turbidity of the water.

AN MER MOD SWF VIR

ZEU computed from the corresponding merged CHL1 products (weighted method), using the following empirical equation:

𝑍𝐸𝑈 = 101.524−0.436𝑦−0.0145𝑦2+0.0186𝑦3_with_y_{= log}

10(CHL1)

Morel et al. 2007

References

Related products

(42)

2.2.21 ZSD

Category Ocean Subsurface Optical PARAMETER _{ZSD, ZSD-DORON}

method Sensor availability ZSD is the Secchi Disk depth (m). It represents the maximum

depth at which a calibrated black and white disk (the so-called Secchi disk) is still visible from the surface. As such, it is a good indication of the maximum depth of underwater vertical visibility. Two algorithms are available. ZSD is computed according to the Morel et al. algorithm, and ZSD-DORON according to Doron et al.

AN MER MOD SWF VIR

(43)

ZSD-DORON (Doron et al.)

ZSD is computed from the corresponding merged CHL1 products (weighted method), using the following empirical equation:

𝑍𝑆𝐷 = 8.5 − 12.6𝑦 + 7.36𝑦2_{− 1.43}_𝑦3_with_y_{= log}

10(CHL1)

The merged ZSD DORON is computed from the corresponding merged fully normalised remote sensing reflectance products at 490 and 555 nm using the DORON method.

Morel et al. 2007 Doron et al.

References

Doron, M., Babin, M., Mangin, A. and O. Fanton d'Andon (2006). Estimation of light penetration, and horizontal and vertical visibility in oceanic and coastal waters from surface reflectance. Journal of Geophysical Research, volume 112, C06003, doi: 10.1029/2006JC004007.

(44)

2.2.22

CHL-CIA

Category Demonstration Biochemical PARAMETER CHL-CIA

Chlorophyll concentration (mg/m3) according to Color Index Algorithm (see below). It is commonly used as a proxy for the biomass of the phytoplankton.

AN MER MOD SWF VIR

CHL-CIA is computed from L3 monthly merged products CHL1 (AVW) and NRRS using the Color Index band ratio algorithm. For chlorophyll concentrations higher than 0.3 mg/m3, CHL-CIA is equal to CHL1. A linear interpolation between the Color Index band ratio algorithm and CHL1 is performed in the range 0.25 to 0.3 mg/m3_.

Hu et al. 2012

References

Hu, C., Lee, Z., and Franz, B., 2012. “ Chlorophyll a algorithms for oligotrophic oceans: A novel approach bases on three-band reflectance difference”. Journal of Geophysical Research, 117, doi:10.1029/2011JC007395.

Other chlorophyll products available in the GlobColour data set: CHL1 (weighted average and GSM), CHL2

(45)

2.2.23 BBPxxx-LOG

Category Demonstration Optical Subsurface PARAMETER BBPxxx-LOG

BBPxxx-LOG is the particulate back-scattering coefficient

(m-1) at wavelengths xxx = 443, 490, 510 and 555 nm. The bbp provide information on the particulate concentration and size distributions.

AN MER MOD SWF VIR

BBPxxx-LOG uses the Non Spectral Algorithm, based on a Neural Network approach. The product is computed from the monthly merged L3 reflectances NRRSxxx.

Loisel et al. 2006

References

H. Loisel, J.-M. Nicolas, A. Sciandra, D. Stramski, and A. Poteau. 2006. Spectral dependency of optical backscattering by marine particles from satellite remote sensing of the global ocean, Journal of Geophysical Research, 111, C09024, doi:10.1029/2005JC003367

The GlobColour BBP product provides the backscattering coefficient at 443 nm determined from the GSM approach. TSM provides the total suspended matter, a product closely linked to bbp, according to the MERIS Neural Network algorithm.

(46)

2.2.24 BBPS-LOG

Category Demonstration Optical Subsurface PARAMETER BBPS-LOG

BBPS-LOG provides the spectral exponent (logarithmic slope) of the particulate back-scattering coefficient. BBPS provides information about the size distribution of particles.

AN MER MOD SWF VIR

BBPS-LOG is computed by linear regression of log(BBPxxx-LOG) at

443, 490, 510 and 555 nm. Loisel et al. 2006

References

See the BBPxxx-LOG info sheet for more information on the product. BBPS-LOG is used to determine the particle size distributions PSD-XXX

(47)

2.2.25 PSD-XXX

Category Demonstration Biochemical PARAMETER PSD-XXX

PSD-XXX (Particle Size Distribution) provide the number density of micro-, nano- and pico-plankton (#/m3) particles in the Ocean, with the following definitions:

 Micro: 20 and 50 µm

 Nano: between 2 and 20 µm

 Pico: between 0.5 and 2 µm

AN MER MOD SWF VIR

Algorithm Reference

The PSD products are computed from the BBPS-LOG according to a semi-analytical formula based on the Mie theory.

Loisel et al. 2006 Kostadinov et al. 2009

References

Kostadinov, T.S., D.A. Siegel, and S. Maritorena. 2009. Retrieval of the Particle Size Distribution from Satellite Ocean Color Observations. Journal of Geophysical Research, VOL. 114, C09015, doi:10.1029/2009JC005303.

Related products

BBPS-LOG is used to determine the particle size distributions PSD-XXX

(48)

2.2.26 POC-SURF

Category Demonstration Biochemical PARAMETER POC-SURF

POC-SURF is the Particulate Organic Carbon (mol/m3_{) at sea}

surface. POC is an important component in the carbon cycle and serves as a primary food sources for aquatic food webs.

AN MER MOD SWF VIR

Algorithm Reference

POC-SURF is computed from monthly merged L3 reflectances NRRS

using a band ratio algorithm. Stramski et al. 2008

References

Stramski. D., R.A. Reynolds, M. Babin, S. Kaczmarek, M.R. Lewis, R. Rottgers, A. Sciandra, M. Stramska, M.S. Twardowski, B.A. Franz, and H. Claustre (2008). Relationships between the surface concentration of particulate organic carbon and optical properties in the eastern South Pacific and eastern Atlantic Oceans, Biogeosci., 5, 171-201.

Related products

The OSS2015 product POC-SURF is equivalent to the GlobColour POC product but it is computed in a different way (from monthly merged reflectances). POC-SURF uses data from MERIS.

(49)

2.2.27 POC-INT

Category Demonstration Biochemical PARAMETER POC-INT

POC-INT is the Particulate Organic Carbon integrated over the vertical water column (mol/m2). POC is an important component in the carbon cycle and serves as a primary food sources for aquatic food webs.

AN MER MOD SWF VIR

Algorithm Reference

POC-INT is computed from POC-SURF and from the Mixed Layer

Depth World Ocean Atlas Climatology, using an empirical formula. Duforêt-Gaurier et al. (2010),

References

Duforêt-Gaurier L., Loisel H., Dessailly D., Nordkvist K., Alvain S., 2010. “Estimates of particulate organic carbon over the euphotic depth from in situ measurements”. Application to satellite data over the global ocean”. Deep-Sea Research, I 57 (2010) 351–367.

Monterey, G., Levitus, S., 1997. Seasonal Variability of Mixed Layer Depth for the World Ocean. NOAA Atlas NESDIS, vol. 14. U.S. Government Print Office, Washington, DC, 96pp.

Related products

(50)

2.2.28 PP-AM, PP-UITZ

Category Demonstration Biochemical PARAMETER PP-AM, PP-UITZ

PP is the primary production (gC/m2_{/day) of the biomass.}

Two algorithms are available PP-AM (Antoine and Morel) and PP-UITZ (Uitz et al.)

PP-AM and PP-UITZ are available only from the GlobColour ftp server.

AN MER MOD SWF VIR

PP-AM

(51)

PP-AM and PP-UITZ are computed from merged monthly L3 products using the respective algorithms.

Antoine and Morel 1996 Uitz et al. 2008

References

Antoine D. and A. Morel (1996). Oceanic primary production: 1. Adaptation of a spectral light-photosynthesis model in view of application to satellite chlorophyll observations. Global Biogeochemical Cycles, 10: 43-55

Uitz J., Y. Huot, F. Bruyant, M. Babin, and H. Claustre (2008). Relating phytoplankton photophysiological properties to community structure on large scales. Limnology & Oceanography, 53: 614\u2013630

(52)

2.2.29 PHYSAT

Category Demonstration Biochemical PARAMETER PHYSAT

Description

PHYSAT is a data set of Phytoplankton functional types.

The freq-xxx products contain the detection frequency of group xxx (0 < freq-xxx < 1):

 0 = group never detected;

 1 = all valid pixels are associated with the group The xxx code refers to the phytoplankton groups:

 Nan: Nanoeucaryotes (group 1)

 Pro: Prochloroccocus (group 2)

 Slc: Synechococcus (group 3)

 Dia: Diatoms (group 4)

 Pha: Phaeocystis-like (group 5)

The GlobalMapOfMonthlyDominantsGroups product contains a map of the dominant (highest detection frequency) group for each month, using the numbering convention defined above. The PHYSAT data set is available from the GlobColour Ftp server only.

References

Ben Mustapha Z., Alvain S., Jamet C, Loisel H. and D. Dessailly. Automatic classification of water leaving radiance anomalies from global SeaWifS imagery : Application to the dectection of phytoplankton groups in open ocean waters, RSE-08794, 2014)

(53)

2.3

The spatial and temporal coverage

2.3.1 The binned products

The GlobColour level-3 binned products have a resolution of 1/24° at the equator (i.e. around 4.63 km) for global products and of 1/96° (i.e. around 1.16 km) for Europe products. They consist of the accumulated data of all merged level 2 products, corresponding to periods of one day (a data-day algorithm is applied), 8 days and a calendar month. 8-days binning periods are continuous, starting from the first day of each calendar year.

The geographical location and extend of each bin is determined by the so-called Integerized Sinusoidal (ISIN) grid. The complete ISIN grid definition is provided in appendix.

In GlobColour binned ISIN products, bins are always written in sequential order, from the southernmost-westernmost bin to the northernmost-easternmost bin. Only valid bins are written in a binned product. Bins with no contributions (i.e. uncovered bins) are not contained in the files as well as the covered bins where no valid data has been found. The spatial resolutions of global and Europe products yield to the following grid characteristics:

Area GLOBAL Europe

Average bin size: 4.63 km 1.16 km

Average bin area: 21.44 km2 1.35 km2

Total number of rows in the grid: 4320

Number of columns at equator: 8640

Number of columns at poles: 3

Total number of bins in the grid: 23,761,676 28,307,867

Table 2-5: Main characteristics of the ISIN grids

2.3.2 The mapped products

The GlobColour level-3 mapped products have a resolution of 1/24°, 0.25° or 1.0° (i.e. respectively around 4.63 km, 28 km and 111 km at the equator) for global products and of 0.015°x0.01° for Europe products. They consist of the flux-conserving resampling of the global level-3 binned products. Daily, 8-days and monthly products are available.

(54)

3

The GlobColour system

3.1

Overall description of the processor

The GlobColour processor is the computation element of the GlobColour processing system. Its function is the transformation of EO level 2 products (or level 3 products) from independent instrument/missions into a single merged level 3 product.

The level-2 products are transformed after the sensor-specific preprocessing to the global and Europe ISIN grids. This binning is separately applied to each level-2 input product for each instrument. Outputs are intermediate spatially binned level-3 products for each instrument, also called level 3 at track level.

The term binning refers to the process of distributing the contributions of the level-2 pixels in satellite coordinates to a fixed level-3 grid using a geographic reference system.

When images of different resolutions are to be accumulated together, if the spatial coverage of each pixel is not taken into account, the importance of the image of the highest resolution are largely predominant over the images of smaller resolutions; this may result in introducing a bias in the final product.

Computing a flux value associated to each pixel may solve that problem. Assuming that the data flux for each input pixel is constant, the resampling problem is actually reduced to the problem of finding the set of pixels overlapping each level-3 bin, and then calculating the relative overlapped area.

This approach not only allows to properly mix data of various resolutions together, it also allows to distribute data properly among different level-3 bins as the input image pixel is usually overlapping several of them. This also makes it possible to produce level-3 data at a higher resolution than the input data with no "holes".

Though very attractive, the major drawback to this method is that it is significantly slower than the usual method; different techniques are being investigated to increase the speed of this approach.

The algorithm implemented in the GlobColour processing chain uses the fast Sutherland-Hodgeman area clipping. For more information on the algorithm used refer to ["A fast flux-conserving resampling algorithm", available at http://skyview.gsfc.nasa.gov/polysamp/]. The same binning algorithm is applied to each kind of input variables. Only the flags taken into account when filtering the data are different. These flags are listed in the next sub-section.

Following this logic, the GlobColour processor is mainly composed of 4 separate modules, namely:

1. a preprocessor module 2. a spatial binning module 3. a merging module

4. a temporal binning module

For each sensor, a pre-processing is foreseen just after extraction of the L2. This preprocessor serves for example in the case of MERIS and wherever requested to transform the L2 normalised water leaving reflectances into fully normalised remote sensing

(55)

reflectance. It could be used also to apply cross calibration LUT to be in position to merge equivalent data.

The complete binning scheme for the production of the GlobColour ocean colour products is a three steps approach comprising spatial binning, data merging and temporal binning as shown in the Figure 3-1.

MERIS L2 SeaWiFS Ocean color L2 MODIS/Aqua Ocean color L2 VIIRS/NPP Ocean Color L2

Preprocessing Preprocessing Preprocessing Preprocessing

(1) Spatial binning (2) Temporal binning in dailymode (3) Merging (4) Temporal binning in periodicmode MERIS L3 track SeaWiFS L3 track MODIS/Aqua L3 track VIIRS/NPP L3 track MERIS L3 daily SeaWiFS L3 daily MODIS/Aqua L3 daily VIIRS/NPP L3 daily Daily merged 8-days Monthly

Level 3 data archive Level 2 data archive

Daily level 3 data cache

(56)

3.2

The preprocessor

A preprocessing function is implemented in the processing chain before applying the binning module. This preprocessing is needed to transform some input data read from the level 2 products into the requested variable. For example, the MERIS normalised water leaving reflectances must be converted into fully normalised remote sensing reflectances.

The preprocessing could be a simple equation or a more complex algorithm using several external auxiliary files.

The following table lists the parameters on which a specific preprocessing is applied. The last column indicates which instrument data is affected.

Acronym Variable description Unit Instruments

NRRS412 Fully normalised remote sensing reflectance at 412 nm sr-1 MERIS (pp1) NRRS443 Fully normalised remote sensing reflectance at 443 nm sr-1 MERIS (pp1) NRRS490 Fully normalised remote sensing reflectance at 490 nm sr-1 MERIS (pp1) NRRS510 Fully normalised remote sensing reflectance at 510 nm sr-1 MERIS (pp1) NRRS560 Fully normalised remote sensing reflectance at 560 nm sr-1 MERIS (pp1) NRRS620 Fully normalised remote sensing reflectance at 620 nm sr-1 MERIS (pp1) NRRS670 Fully normalised remote sensing reflectance at 670 nm sr-1 MERIS (pp1) RRS681 None normalised remote sensing reflectance at 681 nm sr-1 MERIS (pp2) RRS709 None normalised remote sensing reflectance at 709 nm sr-1 MERIS (pp2)

LON, LAT, SZA, SAA, VZA, VAA, PRESSURE, WIND

Geometrical characteristics of the observations MERIS (pp3)

SZA, SAA Geometrical characteristics of the observations MODIS (pp4) - SeaWiFS (pp4) - VIIRS (pp4)

Table 3-1: List of variables with a specific preprocessing (pp1): MERIS fully normalised remote sensing reflectances

The MERIS fully normalised remote sensing relfectances are computed from the normalised water leaving reflectances available in the MERIS level-2 products.

(pp2): MERIS RRS681, RRS709 bands exception

No normalisation is applied to the MERIS RRS681 and RRS709 bands.

(pp3): Geometrical characteristics of the observations

The geometrical characteristics of the observations are provided in the level 2 products for each pixel or every N pixels and frames (e.g. MERIS tie-points). In the latter case, the preprocessing includes the reconstruction of the information for every pixel. For example, in MERIS level 2 products, the geometry observation and some other auxiliary data are stored every 16 pixels and 16 frames. One of the preprocessing tasks is to rebuild the characteristics of each pixel at each frame by bilinear interpolation.

(57)

(pp4): MODIS, SeaWiFS and VIIRS L2 products now provide LAT/LON for each pixel but SZA and SAA are not available so the preprocessor recomputed them from pixel position and date/time.

The following tables list, for each instrument, all variables coming from the preprocessing module, their symbols and their associated validity equation(s).

For all sensors we consider that a pixel is invalid if the absolute value of its sun zenith angle is greater than 70°.

3.2.1 MERIS

For all MERIS products we discard input Level-2 pixels with packed value equal to 0 (except for the flags band).

Acronym Variable Validity equation NRRSxxx

RRSxxx

Fully (or None) normalised remote sensing reflectance at xxx nm

WATER and not (HIGH_GLINT or ABSOA_DUST or PCD_19_WHITECAPS15 [Note 1] or CLOUD or ICE_HAZE

or TOAVI_CSI)

with enlarging HIGH_GLINT, CLOUD, ICE_HAZE and TOAVI_CSI by 2 swath pixels

CHL1 Chlorophyll concentration OC4Me

WATER and not (HIGH_GLINT or ABSOA_DUST or WHITE_SCATTERER

or PCD_15_WHITECAPS15 [Note 1] or CLOUD or ICE_HAZE

or TOAVI_CSI)

with enlarging HIGH_GLINT, CLOUD, ICE_HAZE and TOAVI_CSI by 2 swath pixels

CHL2 Chlorophyll concentration _{C2R NN}

WATER and not (PCD_17 or CLOUD) and not ice from climatology

with enlarging CLOUD by 2 swath pixels

CDM

Coloured dissolved and detrital organic materials absorption coefficient

C2R NN WATER and not (PCD_16 or CLOUD) and not ice from climatology with enlarging CLOUD by 2 swath pixels

TSM

Total suspended matter concentration

C2R NN

WVCS Total water vapor column over clear sky

not (PCD14 or CLOUD or ICE_HAZE or TOAVI_CSI)

with enlarging CLOUD, ICE_HAZE and TOAVI_CSI by 2 swath pixels

T865 Aerosol optical thickness over water

WATER and not (HIGH_GLINT

or PCD_19_WHITECAPS15 [Note 1] or CLOUD or ICE_HAZE

or TOAVI_CSI)

and [ CASE2_S or not (WHITE_SCATTERER or CASE2_ANOM) ]

with enlarging HIGH_GLINT by 2 swath pixels and CLOUD, ICE_HAZE and TOAVI_CSI by 3 swath pixels

A865 Angstrom alpha coefficient over water

T443 Aerosol optical thickness over land

LAND and not (PCD19 or CLOUD or ICE_HAZE or TOAVI_CSI)

and [Note 2]

with enlarging CLOUD, ICE_HAZE and TOAVI_CSI by 3 swath pixels

A443 Angstrom alpha coefficient over land

(58)

ABSD ABSOA_DUST flag statistics

WATER and not (HIGH_GLINT

or PCD_19_WHITECAPS15 [Note 1]) and [ CASE2_S or not (WHITE_SCATTERER

or CASE2_ANOM) ]

Table 3-2: List of parameters and filters applied to the MERIS level 2 data

Note 1: PCD_15_WHITECAPS15 and PCD_19_WHITECAPS15 are respectively

recomputed PCD_15 and PCD_19 flags modified to accept wind speed modulus up to 15 m/s instead of 10 (WHITECAPS is an intermediary internal L2 processing flag raised when the wind speed modulus is greater than the threshold). Detailed definition:

 PCD_15_ WHITECAPS15 = PCD_18 or CASE2ANOM or CASE2Y or (T865 > 0.6) or (wind speed modulus > 15)

 PCD_19_ WHITECAPS15 = PCD_18 or (wind speed modulus > 15)

Note 2: Santer & Vidot aerosol over land algorithm implementation:

 discard pixel if A865 (with scale/offset applied) not in [0, 2.5]

 apply a standard deviation filter on T865 (with scale/offset applied) on a 9x9 box: discard the pixel if it is not in [mean – max(2*stddev, scale), mean + max(2*stddev, scale)], with mean and stddev computed using all T865 valid pixels. The term max(2*stddev, scale) allows to handle homogeneous areas (we don’t want discard all pixels if stddev is very small).

The following parameters are included in the geometrical observation condition:  Sun zenith angle

 Sun azimuth angle  Viewing zenith angle  Viewing azimuth angle

 Mean sea level atmospheric pressure  Surface wind speed

(59)

3.2.2 MODIS/SeaWiFS/VIIRS

Acronym Variable Validity equation NRRSxxx Fully normalised remote

sensing reflectance at xxx nm

not (ATMFAIL or LAND or HILT or HISATZEN or STRAYLIGHT or CLDICE or COCCOLITH or LOWLW or CHLFAIL or CHLWARN

or NAVWARN or MAXAERITER or ATMWARN[Note 1] or NAVFAIL or FILTER or HIGLINT)

CHL1 Chlorophyll concentration CDM

Coloured dissolved and detrital organic materials absorption coefficient at 443 nm

POC Particulate Organic Carbon T865 Aerosol optical thickness

over water

A865 Angstrom alpha coefficient over water

PIC Particulate Inorganic Carbon

not (ATMFAIL or LAND or HISATZEN or STRAYLIGHT or CLDICE or LOWLW

or NAVWARN or ATMWARN[Note 1] or NAVFAIL or FILTER or HIGLINT) NFLH Normalised Fluorescence

Line Height (same as for CHL1) and not(PRODWARN or MODGLINT) PAR Photosynthetically Available

Radiation not(LAND or NAVFAIL or FILTER or HIGLINT) CF Cloud fraction not LAND

Table 3-3: List of parameters and filters applied to the MODIS/SeaWiFS/VIIRS level 2 data

Note 1:ATMWARN is not used for VIIRS

The following parameters are included in the geometrical observation condition:  Sun zenith angle

(60)

3.3

The spatial and temporal binning schemes

The list of steps for the generation of the whole set of GlobColour products is:  step 1: L2 to L3 track on ISIN grid

 step 2: L3 track to L3 daily for each single instrument

 step 3: L3 daily for each single instrument to merged L3 daily  step 4: L3 daily merged to 8days and monthly L3 products

 step 5: L3 daily/8days/monthly mer